Author:
Daizong Ding, Mi Zhang, Hanrui Wang, Xudong Pan, Min Yang ,Xiangnan He
Publication:
In Proceedings of the 30th ACM International Conference on Information and Knowledge Management (CIKM2021),1-5 November 2021, Australia.
Abstract:
Hierarchical community detection, which aims at discovering the hierarchical structure of a graph, attracts increasing attention due to its wide application. However, due to the difficulty of parametrizing the community tree, existing methods mainly rely on heuristic algorithms, which is limited in its low accuracy and unable to handle new observations. As far as we know, how to leverage deep learning techniques for better discovering hierrarchical communities remains almost blank in the existing literature. In this paper, we present the first deep learning framework called ReinCom for hierarchical community detection. To address the challenge of parametrizing the community tree, we propose a novel growing-up process where, at each step, we first partition nodes to the community tree and then adjust the community tree by the partition results. To learn an optimal growing-up process, we propose an embedding agent and a community agent to implement the two sub-steps respectively. Furthermore, we also propose an online learning strategy for new observations on the graph. Empirical results show our proposed model has better modeling effectiveness than the state-of-the-art methods. For example, ReinCom is able to outperform previous works on community detection by 33% relatively in terms of modularity. Besides, with the aid of learned node embeddings, we also devise a graph visualization algorithm which consistently reflects the latent hierarchical structure of a graph.